dorsal/arxiv
View SchemaTask Prototype-Based Knowledge Retrieval for Multi-Task Learning from Partially Annotated Data
| Authors | Youngmin Oh, Hyung-Il Kim, Jung Uk Kim |
|---|---|
| Categories | |
| ArXiv ID | 2601.07474vv1 |
| URL | https://arxiv.org/abs/2601.07474 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Multi-task learning (MTL) is critical in real-world applications such as autonomous driving and robotics, enabling simultaneous handling of diverse tasks. However, obtaining fully annotated data for all tasks is impractical due to labeling costs. Existing methods for partially labeled MTL typically rely on predictions from unlabeled tasks, making it difficult to establish reliable task associations and potentially leading to negative transfer and suboptimal performance. To address these issues, we propose a prototype-based knowledge retrieval framework that achieves robust MTL instead of relying on predictions from unlabeled tasks. Our framework consists of two key components: (1) a task prototype embedding task-specific characteristics and quantifying task associations, and (2) a knowledge retrieval transformer that adaptively refines feature representations based on these associations. To achieve this, we introduce an association knowledge generating (AKG) loss to ensure the task prototype consistently captures task-specific characteristics. Extensive experiments demonstrate the effectiveness of our framework, highlighting its potential for robust multi-task learning, even when only a subset of tasks is annotated.
{
"annotation_id": "7d254cea-c845-4491-8db8-d4ebca0a83ed",
"date_created": "2026-02-17T05:53:12.370000Z",
"date_modified": "2026-02-17T05:53:12.370000Z",
"file_hash": "97afe1ff01f9d185c47f727ace2deeabac22c822d98059948d5a12dfbb4c1ece",
"private": false,
"record": {
"abstract": "Multi-task learning (MTL) is critical in real-world applications such as autonomous driving and robotics, enabling simultaneous handling of diverse tasks. However, obtaining fully annotated data for all tasks is impractical due to labeling costs. Existing methods for partially labeled MTL typically rely on predictions from unlabeled tasks, making it difficult to establish reliable task associations and potentially leading to negative transfer and suboptimal performance. To address these issues, we propose a prototype-based knowledge retrieval framework that achieves robust MTL instead of relying on predictions from unlabeled tasks. Our framework consists of two key components: (1) a task prototype embedding task-specific characteristics and quantifying task associations, and (2) a knowledge retrieval transformer that adaptively refines feature representations based on these associations. To achieve this, we introduce an association knowledge generating (AKG) loss to ensure the task prototype consistently captures task-specific characteristics. Extensive experiments demonstrate the effectiveness of our framework, highlighting its potential for robust multi-task learning, even when only a subset of tasks is annotated.",
"arxiv_id": "2601.07474",
"authors": [
"Youngmin Oh",
"Hyung-Il Kim",
"Jung Uk Kim"
],
"categories": [
"cs.LG",
"cs.AI",
"cs.CV"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "Task Prototype-Based Knowledge Retrieval for Multi-Task Learning from Partially Annotated Data",
"url": "https://arxiv.org/abs/2601.07474",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "76f3d763-ffe4-402a-b62a-653e61cd829a",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}